Theresa May’s humiliation in 2017, losing her majority after calling an election from a position of apparently unassailable dominance, was the moment I started taking election analysis and prediction seriously. This was because I was so wrong about the outcome. At the start of the campaign it was entirely reasonable for me to believe, like everyone else, that the Tories were going to win a huge majority. They were 20 points up in the polls and, a few weeks in, we saw local election results, with big Labour losses, that confirmed those polls were broadly right.
But over the course of the election the data changed. The polling gap narrowed significantly. YouGov launched the first UK MRP (multi-level regression and post-stratification) model which was eerily accurate in predicting the final result at constituency level. By the end of the campaign, polls applying different methodologies were producing wildly different results – everything from a 13 point Tory lead, which would have been a landslide, to a 2 point Labour one. At very least that should have led me to be extremely cautious about the result.
Instead I confidently predicted the Tories would win big right up to the point I heard rumours, a few hours before the exit poll, that private exits were showing a much more favourable result for Labour. It was little solace that, with a few honourable exceptions, the entire pundit class made the same mistake. Like them I had dismissed the growing evidence of a much closer result than anticipated because I saw Jeremy Corbyn as such a flawed candidate. And I also believed the supposed truisms about elections: that campaigns don’t make a difference; that Tory-dominated press coverage would crush a leftwing Labour campaign; that terrorist incidents – we saw two during the campaign – would play best for the party of law and order; and so on.
Of course Labour did still lose the election, winning 55 fewer seats than the Tories, but they came close to making it impossible for Theresa May to form a government. The one she ended up with was incredibly weak and ultimately led to the paralysis and chaos we’ve seen from the Conservatives ever since. But despite the election’s importance its lessons have been routinely ignored. This is partly because some of these are inconvenient to the narratives both Tories and Labour want to pursue now; and partly because the old rules seemed to reassert themselves in 2019 when Corbyn did suffer that expected crushing loss. But this selective memory is leading to strategic mistakes when it comes to next year’s election. So in this post I am going to run through four lessons from 2017 and why they matter today.
Keep reading with a 7-day free trial
Subscribe to Comment is Freed to keep reading this post and get 7 days of free access to the full post archives.